Congratulations to Dr. Subin Erattakulangara for a successful PhD defense.

Thesis Title: Towards Accessible Deep Learning for Upper-Airway Segmentation: Open-Source Tool Development and Application in Voice and Sleep Apnea Studies

Abstract: The human vocal tract is an integral part of our anatomy that enables essential functions such as speaking, singing, and breathing. To study and understand the vocal tract and the conditions affecting it, researchers utilize imaging techniques like magnetic resonance imaging (MRI), which offers detailed, non-invasive views of the vocal tract. The initial step in analyzing this data involves conducting manual segmentation, which is resource-intensive and time-consuming. In his thesis, Subin addresses these challenges in segmenting the upper airway by presenting various segmentation methods for multimodal MRI data that are both data-efficient and robust across different MRI protocols. Additionally, it includes the creation of open-source datasets with expert segmentations for 2D and 3D MRI scans as a contribution to the voice and speech community. The neural networks developed in his work are specifically modified to overcome the scarcity of large datasets in the realm of upper airway MRI. Furthermore, he has released a no-code tool called "AirSeg" to semi-automatically segment the upper airway, enabling experts to perform fast and efficient segmentation.

Whats next: Subin is heading to NY to start a Research Scientist role in Dr. Ricardo Otazo and Dr. Victor Murray's labs at the Memorial Sloan Kettering Cancer Center. We wish him the best in this role.